An estimated 65% of nursing home admissions are suffering from dementia, although little is known about their medical outcomes or the extent to which their dementia is responsible for the use of medical care. Our current study is designed to estimate the prevalence of dementia in a cohort of 2,460 new admissions, aged 65 and older, to a representative sample of 59 nursing homes in Maryland and compare the mortality, morbidity, and medical care use of demented and non-demented residents during the first year in a nursing home. The proposed continuation is designed to examine these outcomes in the same cohort beyond one year, because most persons entering a nursing home with dementia do not return to the community; over 60% stay beyond one year. The proposed continuation has three primary aims: 1) to contrast the rates of mortality, morbidity, and medical care use in demented vs. non-demented admissions during the second and third years of their stay; 2) to describe the changes in the rates of these outcomes from the first year of residence through the third year separately for demented and non-demented residents; and 3) to compare the changes in these outcome rates over time, up to 3 years, between demented and non-demented residents. Currently, new admissions are being identified, enrolled, and followed for one year. Within 23-37 days of admission, information about health, functioning, and psychosocial status is obtained from family members, nursing home staff, and charts. Residents are administered brief tests of cognition and depression. The determination of dementia at admission is made by an expert panel of neurologists and psychiatrists following DSM- IIl-R criteria, after review of this information. Incidence of major medical morbidity (infections, pressure ulcers, falls, febrile episodes) through one year is obtained from medical charts; mortality and use of medical care also are monitored. The proposed continuation will extend this follow-up for an additional two years. This study overcomes many shortcomings of earlier studies of nursing home populations in that: 1) it focuses on health outcomes in demented persons admitted to nursing homes; 2) it follows a first admission cohort to a representative, statewide sample of nursing homes; and 3) residents are assessed directly by research staff, and the designation of dementia is made by an expert clinical panel following standard diagnostic criteria.